Equivariant Geometric Deep Learning and Applications to Life Sciences
Seminar Données et Aléatoire Théorie & Applications
23/01/2025 - 14:00 Sergei Grudinin (CNRS) Salle 106
I will briefly present the history of AI-driven bioinformatics and then demonstrate our recent developments for the prediction and annotation of the 3D structure and dynamics of macromolecules. In particular, I will show architectures that handle arbitrarily shaped volumetric patterns with operations inherently invariant or equivariant to patterns’ positions and orientations in 3D. When benchmarked on diverse volumetric datasets, they demonstrate superior performance over the baselines with significantly reduced parameter counts—up to 1000 times fewer on some benchmarks. I will discuss the applications of our developments to current biological problems and beyond.